The capacity to represent the world in terms of numerically distinct objects is a milestone in early cognitive development and forms the foundation for more complex thought and behavior. Failure to develop this capacity places infants at significant risk for later developmental difficulties. Recent research has revealed important developmental changes in infants' ability to use featural information to individuate objects. However, relatively little is known about the neural mechanisms that underlie these changes, in large part because there are a limited number of non-invasive techniques available to measure brain function in infants. Hence, there is a critical need to identify new methods to specify the relation between behavior and brain function in object processing. One such method is near-infrared spectroscopy (NIRS), an optical imaging technique that uses changes in cerebral blood volume as an indicator of neural activation. Because NIRS is safe and non-invasive, can be used during behavioral tasks, and provides spatial and temporal information about neural activation, it is ideal for infant research of this kind. The research proposed in this application will use NIRS to identity changes in neural activation during object processing tasks. The objective of this application is to identify the neural basis of infants' ability to (a) use shape, size, color, and luminance information to individuate objects and (b) extract shape from a variety of perceptual cues (e.g., coherent motion and contour). The central hypothesis is that infants' capacity to individuate objects using shape, color, and luminance differences will be associated with unique, well-defined patterns of neural activation in higher level object processing areas. In addition, infants' capacity to perceive the shape of an object on the basis of coherent motion and connected contour will be associated with unique patterns of neural activation in extrastriate cortex. This hypothesis has been formulated on the basis of behavioral and neuroimaging data collected in our lab. The rationale for the proposed research is that once we have identified the neural substrates that support featurally-based object representation and individuation in the infant, it will be possible to formulate a developmental model of object processing that focuses on brain-behavior relations.